The controversy surrounding the U.S. federal government’s use of the National Guard as a supposed anti-crime intervention has reignited old debates about crime control, now with higher stakes. Fundamental questions — like why one setting would have a higher crime rate than another — have re-emerged during a time when population geneticists are developing increasingly sophisticated tools to study the roots of social behaviors. And while the computational machinery of today might be unprecedented, larger conversations around our quest to measure criminal propensity sound disappointingly familiar.
September marked the 30th anniversary of a highly controversial conference on the relationship between genetics and criminal behavior that was convened at the Aspen Institute in Queenstown, Maryland. The event, held in 1995, was officially titled “The Meaning and Significance of Research on Genetics and Criminal Behavior” and aimed to bring together experts to probe whether genetic factors might predispose certain people to violent or antisocial behavior. It was a newsworthy event in part because the conference was a second attempt at a gathering on related topics.
Organizers had first proposed a conference in 1992 titled “Genetic Factors in Crime: Findings, Uses and Implications,” scheduling it to take place at the University of Maryland. The gathering was postponed after widespread protests and the withdrawal of financial support from the National Institutes of Health. The university accused the NIH of caving to complaints from organizations like the Congressional Black Caucus and the NAACP, which pushed back against the racial implications of the stated goals. The NIH responded by suggesting that the organizers had distorted the goals of the conference in a manner that negatively impacted the agency’s reputation.
After some thought, the organizers amended the conference’s language to reflect a less problematic goal. But when the event was convened three years later, much of the controversy remained, and it became a touchpoint for debates about scientific inquiry, racial justice, and the persistent specter of biological determinism in criminal justice research.
It is important to note the backdrop of the 1995 conference, which took place one year after two key events: the release of the book “The Bell Curve,” which argued for genetic explanations for group differences in IQ, and the passage of the crime bill, the culmination of years of tough-on-crime policies. The Human Genome Project was also in full gear by 1995, and enthusiasm was brewing that we’d soon reveal the biological bases of diseases and behaviors of many sorts. The conference sought to examine evidence from studies of twins and people who had been adopted regarding the heritability of violent or antisocial behaviors, specific gene variants associated with the regulation of behavior, the physiology of neurotransmitter systems associated with impulsivity and aggression, and other biological bases for deviant behavior.
During the conference, there were protests and heated exchanges, with tensions even spilling into the scientific sessions. On the second day, protesters entered meeting rooms and voiced opinions that challenged speakers directly. The academic response had a long arc but included an edited volume of works that arose from the conference, including many papers that examined the intersection between genetics and crime. The conference spawned a groundswell of interest from scholars, activists, and political actors, reflecting a meaningful connection between science and the public. On the other hand, the original conference’s bombastic framing may have harmed present and future efforts to examine the genetic underpinnings of human behavior, even among scientists like me with no interest in predicting behaviors.
In light of this anniversary, we should take account of what we’ve learned. The search for biological explanations for antisocial behaviors (often correlated with crime) remains intense. And why is that? The same reason that “tough on crime” remains a low-risk political slogan: No one enjoys living in settings where their safety feels compromised. And so there might be nothing essentially vile about the question of why some people might be inclined to commit crimes. But the innocence of basic questions around criminal propensity is only skin deep.
The original conference’s bombastic framing may have harmed present and future efforts to examine the genetic underpinnings of human behavior.
The field of carceral studies has risen to new heights in the last two decades and forces us to rethink everything that we know about crime. Major works have outlined the manner in which U.S. crime control policy was crafted by the (often dubious) decisions of political leaders and how the very basics of crime statistics are muddied with racist motivations rather than genuine concern for public safety. And importantly, they have identified the event horizon between the social causes of crime and the often biological myths attached to them.
Where do fields like genetics — when practiced by well-intentioned, technically sound scientists — come in? Our most sophisticated tools now teach us that the genetic material contributing to any important trait is not housed in single gene variants, as scientists once suspected, but rather is a complex stew made up of the appearance of certain forms of a gene, or alleles, in varied combinations. The number of alleles that are associated with a measurable trait can be condensed, along with their effect sizes, into a single number that captures the statistical association with that trait, known as the polygenic score. But the 30 years since the “crime gene” conference has taught us that shiny statistical instruments won’t solve social problems.
As the social scientists and humanists have taught us, obstacles remain in measuring and defining a subjective trait like “crime.” Just as in 1995, the definition is a political decision — the lines between what is right or wrong are at the whim of whoever wields power or has the largest platform. If we’re trying to target a person’s tendency for violence, we run into the problem of having to explain other non-criminal violent behavior, like that generated by the military, policing, and sports. Another way to put it: The difference between a loyal gang member and a loyal member of an infantry might reside in something less objective than in a countable collection of As, Cs, Gs, and Ts, the stuff of genomic sequences. These conceptual challenges are compounded by technical questions that have complicated our ability to meaningfully apply polygenic scores to complex social traits such as the propensity for crime.
These problems are not limited to the search for genes associated with antisocial behavior. In many ways, the crime gene debate is just a stand-in for the cosmic slop that has characterized our search to find simple genetic answers to all kinds of complex social traits. Sadly, the missteps of the 1995 conference remain unaddressed 30 years later. But there is hope. A new generation of scientists is both equipped with the computational tools to study life in genomic color and is appreciative of the social forces that frustrate our attempts to use flat metrics to answer four-dimensional questions — like why some people commit crime more than others. In the meantime, the conversation remains chock-full of bad assumptions, thin foundations, and dangerous social and political implications.